sign in sign up
<<
Home , Abarelix

Incidence of Skeletal-Related Events in Men With Castration-Resistant Alison Tse Kawai,1 David Martinez,2 Catherine Saltus,1 Zdravko Vassilev,3 Montse Soriano-Gabarrό,4 James A. Kaye1 1RTI Health Solutions, Waltham, MA, United States; 2RTI Health Solutions, Barcelona, Spain; 3Bayer US, Whippany, NJ, United States; 4Bayer AG, Berlin, Germany

• Because Medicare claims data lack biochemical and radiological data, BACKGROUND castration resistance was inferred from the initiation of second-line RESULTS DISCUSSION treatments after surgical or medical castration, identified as follows: • Skeletal-related events (SREs) are common in men with bone Participants Outcomes Figure 1. Incidence Rates of SREs and 95% Confidence Strengths metastases and have negative consequences for patients – Surgical castration (bilateral orchiectomy): Identified using algorithms Intervals (per 100 Person-Months) Among All Person-Time 3 • During the study period, 564,491 men were diagnosed with Table 2. Demographic Characteristics of Patients With and • This analysis used the SEER-Medicare database, which is representative of the with castration-resistant prostate cancer (CRPC), including for orchiectomy that were created for a prior study. and Stratified by Use of Any Bone-Targeted Agent prostate cancer. Of those, 2,234 individuals met the Without SREs During Follow-up (N = 2,234) elderly population in the US and is the largest available source of detailed population- pain, reduced quality of life, and increased risk of death. – Medical castration ( deprivation therapy): Identified with 5.0 protocol-defined criteria for CRPC and other eligibility Number of Patients Number of Patients based medical information on men aged 65 years or older with prostate cancer. • administrations or prescriptions for the following drugs: abarelix, Published data on background rates of SREs in men with criteria for the cohort. Variable With SRE (%) Without SRE (%) 4.0 CRPC under routine medical care are sparse. , , cyproterone, , , (N = 896) (N = 1,338) Limitations estramustine, , , , , leuprolide, Table 1. Demographic and Clinical Characteristics of the 3.0 Race • Second-line treatment (after surgical or medical androgen deprivation therapy) medroxyprogesterone, megestrol, , nilutamide, polyestradiol, Study Cohort (N = 2,234) 2.0 was used to define CRPC because biochemical and radiologic data that directly or . This list was adapted from the American Urological White 769 (85.8) 1,098 (82.1) Variable Number of Patients (%) indicate disease progression despite androgen deprivation therapy are not OBJECTIVE Association Guidelines.4 Black 72 (8.0) 146 (10.9) Incidence rate 1.0 Race available in claims data. Asian 16 (1.8) 30 (2.2) • To estimate the incidence rates of SREs in a cohort of men – Initiation of second-line systemic therapies: Identified with the first White 1,867 (83.6) (per 100 person-months) 0.0 All person-time Person-time before Person-time after • Diagnosis and procedure code algorithms in Medicare claims data were used to with CRPC in the Surveillance, Epidemiology, and End occurrence of administrations or prescriptions of any of the following Hispanic 12 (1.3) 36 (2.7) use of any use of any Black 218 (9.8) identify SREs. Claims codes for radiation do not specify the anatomic target, and Results (SEER)-Medicare linked database treatments: abiraterone, cabazitaxel, docetaxel, , a bone-targeted agent bone-targeted agent Asian 46 (2.1) Other or unknown 27 (3.0) 28 (2.1) codes for fractures may capture fractures due to causes other than pathologic mitoxantrone, or sipuleucel-T. Patients 2,234 1,021 1,539 Hispanic 48 (2.1) Age at cohort entry, years Person-months 23,716 7,429 16,287 processes (including trauma or osteoporosis). a 75.5 (5.9) 77.3 (6.3) Cases 896 309 587 Outcomes Other or unknown 55 (2.5) Mean (SD) • Because of potential bias due to confounding by indication, we did not perform METHODS Distribution • Primary outcome: SREs, defined broadly as fracture, bone surgery, Age at cohort entry, years • The incidence rate of SREs in all person-time was 3.78 per any hypothesis testing or calculate any measures of association comparing rates 65-69 143 (16.0) 154 (11.5) Data Source radiation therapy, or spinal cord compression. Mean (SD) 76.6 (6.2) 100 person-months. before versus after bone-targeted agent use or in patients with versus without 70-74 293 (32.7) 332 (24.8) SREs before cohort entry. • The SEER-Medicare linked database is administered by the • Secondary outcomes: (1) fracture (pathologic or traumatic) and Distribution • The incidence rate was numerically lower after use of any bone- United States (US) National Cancer Institute. (2) pathologic fracture. 65-69 297 (13.3) 75-79 231 (25.8) 364 (27.2) targeted agent (before use: 4.16 per 100 person-months; after 80-84 156 (17.4) 295 (22.0) use: 3.60 per 100 person-months). – Combines data from the SEER program, which collects • Outcomes were identified in Medicare claims data. 70-74 625 (28.0) CONCLUSIONS 75-79 595 (26.6) 85+ 73 (8.1) 193 (14.4) Figure 2. Incidence Rates of SREs and 95% Confidence population-based cancer registry data covering • Patients entered the cohort on the date they were identified as first 1 a Intervals (per 100 Person-Months) Among Patients Without or • In this large cohort of elderly men with CRPC under routine medical care in the approximately 30% of the US population, with data 80-84 451 (20.2) Categories were combined to avoid reporting a count of < 11. receiving a therapy representing a second-line systemic treatment for With an SRE Before Cohort Entry US, SREs were common, with most occurring within 1 year after cohort entry. from Medicare, the US federal health insurance prostate cancer. On this date, follow-up began for the occurrence of 85+ 266 (11.9) • SREs were common, with 40% of the cohort experiencing an program primarily for people aged 65 years or older 5.0 • Incidence rates of SREs were numerically lower following bone-targeted agent an SRE. SRE prior to cohort entryb 1,252 (56.0) SRE during follow-up. use and numerically higher among patients with a prior history of SREs. – Contains detailed information for each primary cancer • 4.0 • Follow-up time for each patient continued until the earliest occurrence First use of bone-targeted agent 1,213 (54.3) No substantial differences were observed in the race and individual, including the initial diagnosis and date prior to cohort entryc • A direct causal interpretation of the difference in rates before versus after bone- of the following: SRE, death, second primary malignancy, discontinuation distribution of patients with and without SREs. 3.0 of death targeted agent use or in patients with versus without SREs before cohort entry is not of Part A or Part B Medicare coverage, enrollment in a health First use of bone-targeted agent 326 (14.6) • after cohort entry Patients with SREs tended to be younger than those 2.0 possible because confounding by indication and other factors cannot be excluded. Study Population maintenance organization, or the end of the study period. without SREs. No use of bone-targeted agents 695 (31.1) Incidence rate 1.0 However, further analysis may address at least some potential confounders. • Men aged ≥ 65 years in the SEER-Medicare database with a Analysis Length of follow-up, months • Mean (SD) time from cohort entry to first SRE was 9 (10.1) (per 100 person-months) 0.0 prostate cancer diagnosis in 2000-2011, who had surgical or months. More than half of the patients (74.3%) developed Patients without any SRE Patients with at least one • Incidence rates of SREs (per 100 person-months) and 95% confidence 10.6 (11.6) medical castration, and who met the protocol-defined Mean (SD) before cohort entry SRE before cohort entry intervals (CIs) were calculated in all eligible person-time following their first SRE within the first year after cohort entry, 18.3% DISCLOSURES criteria for castration resistance and the following: Distribution within the second year after cohort entry, and 7.4% more than Patients 982 1,252 cohort entry (defined as the date of initiation of second-line treatment for Person-months 11,549 12,167 A. Kawai, C. Saltus, D. Martinez, and J. Kaye are employees of RTI Health Solutions, which received < 6 months 960 (43.0) 2 years after cohort entry. – Had no prior malignancy (other than prostate cancer or prostate cancer). Cases 385 511 funding from Bayer HealthCare Pharmaceuticals, Inc. to conduct this study. The contract between nonmelanoma skin cancer) 6 months to 1 year 590 (26.4) RTI Health Solutions and the sponsor includes independent publication rights. RTI International • Incidence rates were also stratified by: > 1 to 1.5 years 317 (14.2) Table 3. Number of Cases of First SREs • The incidence rate of SREs was numerically higher among conducts work for government, public, and private organizations, including pharmaceutical – Were enrolled in Medicare Parts A and B continuously companies. Z. Vassilev and M. Soriano-Gabarro are employees of Bayer. for at least 1 year before initiation of second-line – Person-time before and after first use of any of the following bone- > 1.5 to 2 years 152 (6.8) SRE Cases (% of Total Cohort) patients with a history SRE than patients without such a history (4.20 per 100 person-months versus 3.33 per 100 person-months) This study was conducted using data from NCI’s SEER program of the US and guided by a data use systemic therapy (i.e., cohort entry date) and from date targeted agents: alendronate, denosumab, ibandronate, > 2 years 215 (9.6) Fracture 266 (11.9) pamidronate, risedronate, or zoledronic acid agreement between NCI and RTI Health Solutions. of initial prostate cancer diagnosis until initiation of SD = standard deviation. Bone surgery 22 (1.0) Figure 3. Incidence Rates of SREs and 95% Confidence a second-line systemic therapy • Patients were identified on the basis of whether they used bone- Categories were combined to avoid reporting a count of < 11. Radiation 609 (27.3) Intervals of Fractures and Pathologic Fractures (per 100 targeted agents after the initial prostate cancer diagnosis. b Identified in the Medicare Provider Analysis and Review file, carrier claims, or Person-Months) – Had no evidence of treatment with radium-223 outpatient claims any time between the initial date of prostate cancer diagno- Spinal cord compression 37 (1.7) REFERENCES dichloride prior to the cohort entry date (though none – If no use was identified after the initial diagnosis, all person-time was sis and the cohort entry date. 1.50 c Total 896 (40.1) 1. NCI. http://seer.cancer.gov/about/overview.html. was observed in the data) considered to be “before” first use. Identified any time between the initial date of prostate cancer diagnosis and 1.25 the cohort entry date. Notes: Cases identified using Medicare Provider Analysis and Review file, carrier 2. Heidenreich A, et al. Eur Urol. 2014 Feb;65(2):467-79. • Castration resistant prostate cancer, defined as advanced – If use was identified after the initial diagnosis, we identified whether claims (Physician/Supplier Part B), and outpatient claims. SREs on the date of the 1.00 3. Johannes C, et al. Pharmacoepidemiol Drug Saf. 2011;20(suppl 1):S271-2. initial SRE during follow-up were counted. The sum of the cases of fracture, bone first use was before or after cohort entry. We then categorized person- • The cohort was primarily white (83.6%), and the mean age at 4. Cookson MS, et al. J Urol. 2015 Feb;193(2):491-9. prostate cancer progression despite medical or surgical surgery, radiation, and spinal cord compression is greater than the total number 0.75 time into whether it occurred before or after first use. cohort entry was 76.6 years. castration, is identified in clinical practice using serum of SRE cases, as patients may have had more than one type of SRE on the date 0.50 of the first SRE. – History of SRE before cohort entry Incidence rate levels, prostate-specific antigen • More than half (56.0%) of the cohort had a history of SRE 0.25 measurements, and results of bone imaging studies.2 prior to cohort entry. • The most common types of SREs were radiation (occurring in CONTACT INFORMATION (per 100 person-months) 0.00 27.3% of the cohort) and fracture (11.9% of the cohort). Fractures Pathologic fractures • Bone-targeted agent use in the cohort was common, with Alison Tse Kawai, ScD Patients 2,234 2,234 Senior Research Epidemiologist initial use before cohort entry in 54.3% of patients and initial • Spinal cord compression and bone surgery were less Person-months 31,257 33,635 use after cohort entry in 14.6% of patients. common (each in < 2% of the cohort). RTI Health Solutions Cases 363 176 E-mail: [email protected]

The power of knowledge. The value of understanding. Presented at: Palliative and Supportive Care in Oncology Symposium; November 16-17, 2018; San Diego, CA, United States